Cytochromes P450 1B1, 1A1 & 1A2 (CYP1B1, CYP1A1, and CYP1A2) are responsible for both detoxifying and metabolically activating innumerable polycyclic aromatic hydrocarbons (PAHs), nitrosamines, and A/-heterocyclics present in combustion processes including cigarette smoke. The PAH- inducible CYP1B1/1A1/1A2 genes are up-regulated by the aromatic hydrocarbon receptor (AHR). Many in vitro, cell culture and animal studies have shown that high CYP1 enzyme levels and AHR high-affinity are correlated with increased genotoxicity caused by PAHs; recent studies with knockout mice, however, show that, whereas high CYP1B1 activity causes metabolic activation, CYP1A1 and CYP1A2 are far more important in detoxication than metabolic activation. Data in humans have been inconclusive, perhaps because many are assuming that all three CYP1 enzymes cause only higher cancer risk. We have a large cohort of head-and-neck squamous-cell carcinoma (HNSCC) patients with a history of one to 40 cigarette pack-years ("highly sensitive," HS) and heavy smokers with >80 pack-years having no cancer ("highly resistant," HR)?strongly suggesting a genetic component. These two extremes will be studied, using the extreme discordant phenotype (EDP) approach. A systematic search (SNP-discovery followed by SNP- typing) for haplotypes of the human,CYP1B1 and AHR genes is now possible; we have completed such a study of the CYP1A1_1A2 locus (39.6 kb) and discovered 85 SNPs, 49 of which were not yet in any database. Our hypothesis is: Specific haplotypes involving one or more of these four genes leading to high CYP1B1 and low CYP1A1/1A2 activities are associated with increased risk of HNSCC cancer in smokers. In the 3 years of this proposed project, we therefore will: [a] carry out whatever SNP-discovery and SNP-typing that still needs to be done, from six major geographically-isolated subgroups; we will collect blood and prepare DMA from 200 HS HNSCC patients and 200 HR non-cancer heavy smokers in our cohort; and [b] examine the association between haplotypes and risk of HNSCC by performing SNP-typing of the CYP1B1, CYP1A1, CYP1A2 and AHR genes in the 200 HS versus the 200 HR individuals. Establishing important phenotype-genotype associations between HNSCC and these four genes would provide the first unequivocal data that humans are similar to laboratory animals regarding cancer and one or more of these genes. [unreadable] [unreadable] [unreadable]